|Home | About | Journals | Submit | Contact Us | Français|
β-adrenergic receptors (β-ARs) modulate cardiotoxicity/cardioprotection through crosstalk with multiple signaling pathways. We have previously shown that β2-ARs are cardioprotective during exposure to oxidative stress induced by doxorubicin (DOX). DOX cardiotoxicity is mediated in part through a Ca2+-dependent opening of the mitochondrial permeability transition (MPT), however the signals linking a cell surface receptor like the β2-AR to regulators of mitochondrial function are not clear. The objective of this study was to assess mechanisms of crosstalk between β2-ARs and mitochondrial cell death pathways.
DOX administered to WT mice resulted in no acute mortality, however 85% of β2-/- mice died within 30 min. Several pro- and anti-survival pathways were altered. The pro-survival kinase, εPKC, was decreased by 64% in β2-/- after DOX vs WT (p<0.01); the εPKC activator ψεRACK partially rescued these mice (47% reduction in mortality). Activity of the pro-survival kinase Akt decreased by 76% in β2-/- after DOX vs WT (p<0.01). The α1-antagonist prazosin restored Akt activity to normal and also partially reversed the mortality (45%). Deletion of the β2-AR increased rate of Ca2+ release by 75% and peak [Ca2+]i by 20% respectively in isolated cardiomyocytes; the Ca2+ channel blocker verapamil also partially rescued the β2-/- (26%). Mitochondrial architecture was disrupted and complex I and II activities decreased by 40.9% and 34.6% respectively after DOX only in β2-/-. The MPT blocker cyclosporine reduced DOX mortality by 41% and prazosin plus cyclosporine acted synergistically to decrease mortality by 85%.
β2-ARs activate pro-survival kinases and attenuate mitochondrial dysfunction during oxidative stress; absence of β2-ARs enhances cardiotoxicity via negative regulation of survival kinases and enhancement of intracellular Ca2+, thus predisposing the mitochondria to opening of the MPT.
The two major cardiac β-adrenergic receptor (β-AR) subtypes, β1 and ß2, modulate cardiac signaling through both parallel and opposing pathways. Over the past decade, our understanding of the function of these key receptors has increased beyond their classical role in regulating inotropy and chronotropy, and now includes the complex regulation of cardiac remodeling. Stimulation of β1-ARs can induce apoptosis through PKA-dependent  and independent (e.g. CaMKII) pathways . In contrast, stimulation of β2-ARs can be anti-apoptotic, mediated through Gi, Gβγ, phosphatidylinositol 3'-kinase and Akt . In addition, β2-ARs can also signal through ERK via a GRK5/6-β-arrestin-dependent pathway, independent of G protein coupling .
We have previously shown that β1-ARs play a cardiotoxic and that β2-ARs play a cardioprotective role in oxidative stress, using as a model the anthracycline anti-cancer drug doxorubicin (DOX) [5, 6]. β2-AR knockout (β2-/-) mice receiving a single therapeutic-level dose of DOX (which has no acute effect on WT or β1-/- mice) show markedly enhanced cardiotoxicity, with ECG, blood pressure and contractility changes within 2 min, and death within 30 min. Differential activation of MAPK isoforms was observed: p38 activity increased 20-fold only in the β2-/- and p38 inhibition partly rescued this phenotype. However, the exact mechanisms mediating the extremely rapid demise of the β2-/- mice and the additional pathways activated by β2-ARs to protect against anthracycline-induced oxidative stress are unknown.
In addition to the MAPKs, several other kinases play important roles in modulating cardiotoxicity/cardioprotection, including protein kinase C (PKC) and Akt. Of the 11 isozymes of PKC identified, opposing cardioprotective actions have been attributed to the novel subfamily members εPKC and δPKC. εPKC attenuates, whereas δPKC increases, ischemia/reperfusion injury . Similar roles have been observed in both ethanol  and reactive oxygen species-induced cardioprotection , and may explain some of the cardioprotective effects of diazoxide . Akt, a key regulator of cardiac growth and metabolism, also plays a central role in cardioprotection. Activation of Akt inhibits hypoxia-induced myocyte apoptosis  and plays a role in both ischemic preconditioning and postconditioning . In DOX cardiotoxicity, Akt activation protects against myocyte apoptosis , attenuates heart failure  and may mediate the protective effects of dexrazoxane . The interrelationship between these pro-survival kinases and β2-ARs in mediating protection against DOX cardiotoxicity is unknown.
A central intracellular target of DOX toxicity, and a point of convergence of signaling by the above kinases, is the mitochondria. The extremely rapid demise of the β2-/- mice when exposed to DOX makes apoptosis an unlikely mechanism, and we have found no evidence of either apoptosis or ischemia. Rapid cell death in this time frame can be mediated by opening of the mitochondrial permeability transition (MPT) pore in the inner mitochondrial membrane, causing reversal of the F0-F1 ATPase and collapse of the mitochondrial membrane potential . Previous evidence implicates mitochondrial dysfunction in DOX toxicity. DOX is redox cycled through mitochondrial complex I NADH dehydrogenase , increasing the formation of free radical intermediates. Mitochondria are also targets for the [Ca2+]i dysregulation and bioenergetic failure that are hallmarks of DOX cardiotoxicity . However, how signaling mediated through a cell surface G-protein coupled receptor (GPCR), such as the β2-AR, can so profoundly affect these mitochondrial cell death pathways is unknown.
In this study, we sought to determine the mechanisms by which a cell surface GPCR like the β2-AR can crosstalk with mitochondrial cell death pathways to mediate cardiotoxicity/cardioprotection. We used DOX cardiotoxicity as a well established and highly reproducible model of oxidative stress. We hypothesized that absence of β2-AR signaling would lead to subtle alterations in pro-survival kinase signaling and in [Ca2+]i regulation. These alterations would be below the threshold for altering baseline cardiac physiology. However, when exposed to an oxidative stressor such as DOX, these alterations would combine to predispose to opening of the MPT and subsequent cell demise.
DOX was used as a model of oxidative stress and has been well established in the mouse [6, 18, 19]. 3 month old male WT and β2-/- mice, both on a congenic FVB background, were injected with a single dose of DOX (NovaPlus, Bedford, OH) (200–300 μl) via the dorsal tail vein. Three different doses were administered in order to maximize the effect of the inhibitors tested: 8 mg/kg (LD50 dose), 10 mg/kg and 15 mg/kg. These doses are within the normal range of dosing for patients with malignancies, based on human equivalent dosing models [20, 21]. Mortality was assessed at any time during the first 72 hours; however, all the mice that died did so at 20-30 min after DOX. Animal protocols were approved by the Stanford University Institutional Animal Care and Use Committee and conformed with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996).
Mice were administered PKC isozyme-specific regulatory peptides as previously described . The εPKC activator peptide ψεRACK (20 nmol in 150 μl) and the δPKC inhibitor peptide δV1-1 (20 nmol in 150 μl) were each given i.p. 10 min before DOX. The α1-AR blocker prazosin (0.1 mg/kg) was administered i.v. 10 min before DOX . The L-type Ca2+ channel blocker verapamil (25 mg/kg) was given i.v. 30 min before DOX . Cyclosporine (10 mg/kg), a blocker of the MPT, but also of the phosphatase calcineurin, was given i.v. 30 min before DOX . FK-506 (0.5 mg/kg), a blocker of calcineurin but not the MPT  was given i.p. 30 min before DOX. The dose of DOX used in each case was the highest dose at which mortality could be decreased by the inhibitor, i.e. If an inhibitor decreased mortality at 10 mg/kg it was not tested at the lower 8 mg/kg (LD50) dose.
To isolate particulate and soluble fractions, whole-tissue lysates from mouse hearts were prepared as previously described . The soluble and particulate fractions were separated by high-speed centrifugation. εPKC and δPKC levels and translocation were determined by SDS-PAGE, followed by Western blot analysis with anti-PKC, anti-εPKC and anti-δPKC antibodies (Santa Cruz Biotechnology, Santa Cruz, CA), GAPDH was used as loading control (Advance Immnunochemical, Long Beach, CA).
Akt levels and phosphorylation were measured by Western blotting using Akt and Phospho-Akt (Ser473) antibodies (Cell Signaling Technology, Danvers, MA). Immobilized Akt monoclonal antibodies (Cell Signaling Technology, Danvers, MA) were used to immunoprecipitate Akt from lysates, and an in vitro kinase assay was performed using GSK-3 Fusion Protein as a substrate. Phosphorylation of GSK-3 was measured by Western blotting, using Phospho-GSK-3α/β (Ser21/9) antibody.
Hearts from WT and β2-/- mice treated with and without DOX 15 mg/kg were collected 20 min after DOX; hearts from mice who survived 24 h after 15 mg/kg DOX treatment in the presence of a p38 inhibitor were also used for electron microscopy studies. Samples were processed at the Stanford Cell Science Imaging Facility. Heart samples were fixed in 2% glutaraldehyde (EMS, Hatfield, PA) and 4% formaldehyde (EMS). Samples were observed in the JEOL 1230 TEM at 80kV and photos taken using a Gatan Multiscan 791 digital camera (Gatan, Pleasanton, CA).
The activities of complex I and complex II in WT and β2-/- mice without and with DOX 15 mg/kg 20 min post treatment were assessed spectrophotometrically as previously reported .
Adult ventricular myocytes were isolated from β2-/- mice and WT littermates based on previously published protocols, with modifications [29, 30]. Experiments were performed with freshly isolated myocytes resuspended in a HEPES-buffered solution.
Ca2+ transients were evaluated with a video-based sarcomere spacing acquisition system (SarcLen, IonOptix, Milton, MA). Rod-shaped myocytes with clear striation patterns and quiescent when unstimulated were chosen. Cells were electrically stimulated with suprathreshold voltage at 0.5 Hz and superfused with a HEPES-buffered solution. Background-corrected Fura 2 ratios were collected at 510 nm [31, 32].
Data are expressed as mean ± SD. Unpaired t tests were used for comparisons between 2 groups, and ANOVA with Fisher's test was used for differences among >2 groups. Significance was attained with a p<0.05.
To gain insight into whether β2-ARs mediate their protective effects trough crosstalk with specific PKC isozymes, levels of εPKC and δPKC and their translocation from soluble to particulate fractions were assessed at baseline and 20 min. after DOX (15 mg/kg) administration to WT and β2-/-mice.
At baseline there was no difference in εPKC levels in either soluble or particulate fraction between WT and β2-/-, however baseline δPKC was increased slightly (16.0±4.6%) in the particulate fraction in β2-/- compared to WT (Figure 1A). 20 min after DOX, however, εPKC significantly decreased in both particulate and soluble fractions by 67.3±12.3% and 64.0±9.9% respectively in β2-/- whereas δPKC increased by 22±5% in the particulate fraction compared to WT (Figure 1B).
To examine the consequences of these PKC isozyme-specific alterations in mediating enhanced DOX cardiotoxicity in the β2-/-, the δPKC inhibitor (δV1-1) and the εPKC activator (ψεRACK) were each given prior to DOX. When DOX was given at the LD50 concentration of 8 mg/kg, the εPKC activator ψεRACK decreased mortality by 47%, although it failed to rescue the acute cardiotoxicity when DOX was given at the higher doses of 15 mg/kg or 10 mg/kg, which induce 100% and 85% mortality in the ß2-/-, respectively. In contrast, there was no improvement in survival when mice were given the δPKC inhibitor δV1-1 (Figure 1C).
To determine whether β2-ARs crosstalk with other pro-survival kinases in the regulation of cardiotoxicity/cardioprotection, Akt levels, phosphorylation and activity were assessed at baseline and 20 min. after 15 mg/kg DOX was given to WT and β2-/- mice. There was no difference between WT and β2-/- mice in Akt levels at baseline or after DOX (Figure 2A). Of interest, phosphorylation of Akt was lower in the β2-/- compared to WT at baseline and only WT mice showed a decrease in phosphorylation of Akt after DOX, achieving levels similar to those in DOX β2-/- treated mice (Figure 2B). At baseline, there was no difference in Akt activity in β2-/- compared with WT. However, after DOX, Akt activity decreased by 76% in β2-/- but not in WT (Figure 2C). As previous studies have shown that α1-AR signaling can alter regulation of Akt (usually, but not always positively) [33, 34], we administered the α1-AR blocker prazosin (0.1 mg/kg) 10 min before DOX and measured Akt activity. In the β2-/-, prazosin restored Akt activity to normal as early as 20 min and for as long as 24 h after DOX. Prazosin alone did not alter Akt activity from baseline levels (Fig 2D). In β2-/- mice prazosin reduced mortality from 85% to 47% after DOX (10 mg/kg) (Figure 2E).
To determine the role of mitochondrial dysfunction in acute DOX cardiotoxicity in the β2-/-, we examined both structural and functional alterations. Because of the rapidity with which β2-/- mice die after receiving DOX, neither light nor electron microscopy (EM) demonstrated any ultrastructural changes. There were also no signs on EM typical of acute ischemia, such as coronary endothelial swelling. This mortality is totally binary: even when given the LD50 dose of DOX, those β2-/- mice that die do so within 30 min; none survive longer. Those that survive do not manifest late acute toxicity up to 72 hrs after drug administration. We have previously shown that we can partially rescue these mice with the p38 inhibitor SB203580, which allowed us to repeat these pathologic examinations at 24 h, a time when non-treated β2-/- do not survive. LV myocardium from β2-/- mice treated with 15 mg/kg DOX for 24h in the presence of a p38 inhibitor exhibited mitochondria with disrupted, slit-like vacuolization of the cristae. In contrast, none of these changes were observed in WT mice treated with DOX (Figure 3).
Mitochondrial electron transport chain function was evaluated next. At baseline, there were no differences in complex I and II activities between β2-/- and WT. However, after DOX, complex I activity decreased by 41% and complex II activity by 35% in the β2-/-. There was no change in these activities in the WT (Figure 4).
Doxorubicin is known to decreases mitochondrial Ca2+ loading capacity and β1- and β2-ARs regulate [Ca2+]i through both parallel and opposing mechanisms. β1-ARs couple to the stimulatory G protein, Gs, whereas β2-ARs can couple to both Gs and Gi . We have previously shown that resting cardiac contractile function is unaltered in β2-/- mice , however, subtle changes in Ca2+ transients could be subthreshold to affect resting function, but critical in the setting of cardiac oxidative stress. In agreement with our hypothesis, cardiomyocytes isolated from β2-/- mice showed increases in the rate of Ca2+ release (75%) and in peak [Ca2+]i (20%) compared to WT (Figure 5A-C). These changes were not associated with an increase in contractility in β2-/- myocytes as measured by intrasarcomeric shortening (Figure 5D). Although we did not see statistically significant changes, there was a trend toward higher velocity of contraction and relaxation in the β2-/- compared to WT (data not shown). Confirming the role of altered intracellular Ca2+, blocking the L-type Ca2+ channel with verapamil (25 mg/kg) decreased mortality by 26% in β2-/-mice treated with 10 mg/kg DOX (Figure 5E).
Finally, to determine if opening of the MPT was involved in this enhanced cardiotoxicity in the β2-/-, cyclosporine (10 mg/kg), a blocker of the MPT was administered 30 min. prior to 10 mg/kg DOX. Cyclosporine reduced DOX mortality from 85% to 50%. Since cyclosporine is also an inhibitor of the phosphatase calcineurin, we administered FK506, a blocker of calcineurin but not of the MPT, to determine whether this effect was related to calcineurin signaling. As expected if the effect was solely on the MPT, FK506 did not result in a reduction in DOX mortality (Figure 6). One traditional marker of MPT opening, cytochrome C release, was not detected in either WT mice or β2-/- mice treated with DOX, however the rapid time to demise of these mice would likely be too soon for activation of the cytochrome C-intrinsic apoptotic pathway . At baseline, β2-/- mitochondria were not more sensitive to Ca2+-mediated swelling than WT mitochondria (data not shown). Finally, to determine whether alterations in Akt and opening of the MPT were independent or dependent pathways, we administered prazosin and cyclosporine in combination 30 min before DOX. There was a synergistic marked decrease in DOX mortality from 85% to 12.5% (Figure 6).
Our results suggest that β2-ARs mediate their cardioprotective effects in oxidative stress in part through crosstalk with mitochondrial cell death pathways. Anthracycline cardiotoxicity was used as model of oxidative stress to dissect the mechanisms by which a cell surface GPCR receptor such as the β2-AR can influence mitochondrial signaling. Although the exact mechanism of DOX cardiotoxicity is still incompletely defined, the mitochondria are a prime target for DOX action. Cardiomyocytes may be particularly susceptible to DOX toxicity due to their large number of mitochondria; redox cycling of DOX by complex I NADH dehydrogenase in mitochondria further increases the generation of free radicals and further increases cell damage .
We present evidence that β2-ARs mediate their cardioprotective effects in part through crosstalk with the pro-survival kinases εPKC and Akt, which themselves regulated mitochondrial function. The cardioprotective role of εPKC has previously been demonstrated in models of ischemia/reperfusion and preconditioning [40, 41]. Activation of εPKC before ischemia protects mitochondrial function and diminishes apoptosis and oncosis; its cardioprotective mechanisms involve opening of mitochondrial ATP-sensitive K+ channels, prevention of MPT opening and regulation of cytochrome c oxidase activity . In contrast, δPKC has been shown to mediate cardiotoxicity , δPKC is required for DOX-induced apoptosis in Jurkat cells, mediated through sequential activation of caspase-2, δPKC and JNK . Here we demonstrate the role of εPKC and δPKC isozymes in β2-AR mediation of DOX cardiotoxicity. Although normal at baseline, εPKC is markedly reduced in β2-/- mice after DOX in both particulate and soluble fraction. The role of εPKC in this enhanced toxicity was further demonstrated by administration of the εPKC activator, ψεRACK, which partially rescues this acute toxicity. In contrast, although δPKC was slightly increased at baseline in the β2-/-, its inhibition did not influence the cardiotoxicity. β2-AR signaling has been shown to crosstalk with εPKC in regulating ERK activity, although in vitro the effect of εPKC appears to limit β2-AR mediated cardioprotection . The discrepancy between this in vitro study and the current in vivo study may be due to the different stresses used, the species, or in differences between isolated myocytes and intact myocardium.
Our results also provide evidence of β2-AR crosstalk with Akt in mediating cardioprotection . The role of Akt in DOX cardiotoxicity has been explored previously, with some studies suggesting that DOX induces the activation of both caspase-3 and Akt , whereas others show that DOX decreases Akt . Akt activation may be enhanced by DOX-induced ROS generation , and may induce cardioprotection by attenuating cardiomyocyte apoptosis through phosphorylation of GSK-3β . If Akt is a protective mechanism induced in response to DOX, then in the absence of the β2-AR, the marked disregulation of Akt (where activity falls only in the β2-/-) would participate in the enhanced toxicity. In support of this mechanism, the α1-antagonist prazosin restores Akt activity to normal levels, and partially rescues the β2-/- mice. The effect that we observed of α1-AR blockade with prazosin on Akt activity is in agreement with findings by Ballou et al . They have shown, using Rat-1 fibroblasts stably expressing the human α1A-AR, that treatment of these cells with phenylephrine did not activate PI3-kinase or Akt. Furthermore, phenylephrine blocked the insulin-like growth factor-I-induced activation of PI3-kinase and the phosphorylation and activation of Akt-1. In their study, the effect of phenylephrine was not confined to signaling pathways that include insulin receptor substrate-1, as the α1-AR agonist also inhibited the platelet-derived growth factor-induced activation of PI3-kinase and Akt-1. Thus, the α1-AR negatively regulated the PI3-kinase/Akt pathway, resulting in enhanced cell death following apoptotic insult. Taken together with the current study, these results implicate altered regulation of Akt as one of the mechanisms for enhanced cardiotoxicity in the β2-/-.
In addition to these alterations in pro-survival kinases, we hypothesized that altered Ca2+ transients would also provide a link between cell surface β2-ARs and intracellular mitochondrial signaling. Altered Ca2+ uptake has previously been shown to increase mitochondrial ROS production  and DOX is known to decrease mitochondrial Ca2+ loading capacity. . In the absence of β2-AR signaling, peak [Ca2+]i and the rate of Ca2+ release were increased, potentially due to the loss of inhibitory signaling through Gi. Devic et al. have shown increases in isoprotenerol-stimulated contraction rate in β1-/- neonatal cardiomyocytes treated with pertussis toxin . Furthermore, Kuschel et al. have shown that inhibition of Gi with pertussis toxin permits a full phospholamban phosphorylation and a de novo relaxant effect following β2-AR stimulation, converting the traditionally localized β2-AR signaling to a global signaling mode similar to that of β1-AR. The current study would be the first demonstration of an alteration in [Ca2+]i associated with the absence of β2-AR signaling. We have previously shown that baseline cardiovascular function is not altered in the β2-/- mouse , and here show that intrasarcomeric shortening is also not altered (Figure 5D), suggesting that these alterations in Ca2+ are subthreshold at rest. At baseline, β2-/- mitochondria are not inherently more sensitive to Ca2+-mediated swelling than WT mitochondria. However, in vivo when β2-/- mice are stressed with DOX, the increase in [Ca2+]i, combined with alterations in εPKC and Akt signaling, disrupt normal mitochondrial Ca2+ buffering capacity, predisposing to MPT opening and rapid demise. In support of this mechanism, both the L-type Ca2+ channel blocker verapamil and the MPT blocker cyclosporine rescue the enhanced toxicity. Opening of the MPT occurred without cytochrome c release in β2-/- mice. Although cytochrome c release can be observed as early as one hour after ischemia-reperfusion injury , the more rapid demise of the β2-/- after DOX may be too early for activation of the cytochrome c-intrinsic apoptotic pathway. In addition, MPT opening is not always a sufficient condition, in itself, to effect cytochrome c release . We attempted to confirm MPT opening in vitro by treating myocytes isolated from WT and ß2-/- mice with DOX. Unfortunately, DOX autofluorescence interfered with the ability to detect calcein release (a measure of MPT opening) or with the ability to measure mitochondrial membrane potential using TMRM.
The convergence of these pro-survival kinase pathways with intracellular Ca2+ regulation appears to be at the level of the MPT. The multi-component structure of the MPT integrates information on mitochondrial energy metabolism, oxidative stress and Ca2+ load to determine whether or not to activate cell death pathways, and to what extent . Akt's cardioprotective actions are mediated in part through inhibition of GSK-3β's actions on the MPT complex . We have shown that restoring Akt activity with prazosin combined with inhibition of the MPT with cyclosporine had a highly synergistic effect against DOX toxicity in the β2-/- (Fig. 4C), nearly eliminating the enhanced toxicity. Further evidence that mitochondrial dysfunction plays a central role in DOX cardiotoxicity of the β2-/- is evidenced by decreases in complex I and II activities and mitochondrial ultrastructural abnormalities seen only in the β2-/- that survived more than 24 h after the highest dose of DOX combined with p38 inhibition and not in the WT . Although the mitochondria in hearts from β2-/- mice after DOX treatment are not typically swollen, MPT opening is not necessarily and readily followed by matrix swelling in intact cells, (Dr. Fabio Di Lisa, University of Padova, Italy, personal communication).
In conclusion, we have demonstrated a link between a cell surface GPCR receptor and mitochondrial cell death pathway signaling, mediated via pro-survival kinases including εPKC and Akt, and by altered regulation of intracellular Ca2+ transients. When β2-AR signaling is ablated and this is coupled with the oxidative stress induced by DOX, these signaling alterations combine to dramatically increase cardiotoxicity by predisposing to opening of the MPT pore. A similar cardioprotective role for β2-AR signaling has been invoked in the suggestion that a ß2-AR agonist, combined with a β1-AR antagonist, may be superior to a non-specific β-AR antagonist in the treatment of heart failure . The implications of our data for the clinical use of subtype-specific or non-specific β-AR blockers remains to be determined, however, the extension of these previously observed subtype-specific effects to another model of cardiac stress (doxorubicin oxidative stress) is an important step.
We are grateful to Dr. Joel Karliner, Dr. Gary Cecchini, Dr. Elena Maklashina and Dr. Conrad Alano at UCSF/San Francisco Veteran's Hospital and Dr. Hannes Vogel at Stanford for their advice.
This work was supported by the National Institutes of Health Grant HL061535 to D.B.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of Interest